HW8

library(vegabrite)
library(jsonlite)
library(tidyr)
library(dplyr)

Attaching package: 'dplyr'

The following objects are masked from 'package:stats':

    filter, lag

The following objects are masked from 'package:base':

    intersect, setdiff, setequal, union

Exercise 1

supdem <- jsonlite::fromJSON(
  "https://calvin-data304.netlify.app/data/swd-lets-practice-ex-2-3.json"
  )

supdem_long <- supdem %>%
  pivot_longer(
    cols = c(capacity, demand), 
    names_to = "category", 
    values_to = "value"
    ) %>%
  mutate(
    category = factor(
      category, 
      levels = c("demand", "capacity")
      )
    )

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_long),
  mark = "bar",
  encoding = list(
    x = list(
      field = "date",
      type = "ordinal",
      axis = list(title = NULL)
    ),
    y = list(
      field = "value",
      type = "quantitative",
      axis = list(title = NULL)
    ),
    color = list(
      field = "category",
      type = "nominal",
      legend = list(title = "Category"),
      sort = list("demand", "capacity") 
    ),
    xOffset = list(
      field = "category",
      sort = list("demand", "capacity") 
    )
  )
))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_long),
  mark = "bar",
  encoding = list(
    x = list(
      field = "date", 
      type = "ordinal", 
      axis = list(title = NULL)
      ),
    y = list(
      field = "value", 
      type = "quantitative", 
      axis = list(title = NULL)
      ),
    color = list(
      field = "category", 
      type = "nominal", 
      legend = list(title = "Category"), 
      sort = list("demand", "capacity")
    ),
    order = list(
      field = "category",
      type = "nominal"
    ),
    size = list(
      field = "category", 
      type = "nominal",
      scale = list(
        domain = c("demand", "capacity"),
        range = c(8, 16)
      ),
      legend = NULL
    )
  )
))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_long),
  layer = list(
    list(
      mark = "line",
      encoding = list(
        x = list(
          field = "date",
          type = "ordinal",
          axis = list(title = NULL)
        ),
        y = list(
          field = "value",
          type = "quantitative",
          axis = list(title = NULL)
        ),
        detail = list(field = "date"),  
        color = list(value = "gray")
      )
    ),
    list(
      mark = "point",
      encoding = list(
        x = list(
          field = "date",
          type = "ordinal"
          ),
        y = list(
          field = "value", 
          type = "quantitative"
          ),
        color = list(
          field = "category", 
          type = "nominal", 
          legend = list(title = "Category"),
          sort = list("demand", "capacity")
          )
      )
    )
  )
))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_long),
  mark = "line",
  encoding = list(
    x = list(
      field = "date",
      type = "ordinal",
      axis = list(title = NULL)
    ),
    y = list(
      field = "value",
      type = "quantitative",
      axis = list(title = NULL)
    ),
    color = list(
      field = "category",
      type = "nominal",
      legend = list(title = "Category"),
      sort = list("demand", "capacity")
    )
  )
))

supdem_cap <- supdem %>%
  mutate(
    unmet_demand = demand - capacity
  )

supdem_long_cap <- supdem_cap %>%
  pivot_longer(
    cols = c(capacity, unmet_demand),
    names_to = "category",
    values_to = "value"
  ) %>%
  mutate(
    category = factor(category, levels = c("capacity", "unmet_demand")) 
  )

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_long_cap),
  mark = "bar",
  encoding = list(
    x = list(
      field = "date",
      type = "ordinal",
      axis = list(title = NULL)
    ),
    y = list(
      field = "value",
      type = "quantitative",
      axis = list(title = NULL)
    ),
    color = list(
      field = "category", 
      type = "nominal", 
      legend = list(title = "Category"), 
      sort = list("unmet_demand", "capacity")
    ),
    order = list(
      field = "category",
      type = "nominal"
    )
  )
))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = supdem_cap),
  layer = list(
    list(
      mark = "line",
      encoding = list(
        x = list(
          field = "date",
          type = "ordinal",
          axis = list(title = NULL)
        ),
        y = list(
          field = "unmet_demand",
          type = "quantitative",
          axis = list(title = NULL)
        )
      )
    ),
    list(
      mark = list(
        type = "point", 
        filled = TRUE
        ),
      encoding = list(
        x = list(
          field = "date", 
          type = "ordinal"
          ),
        y = list(
          field = "unmet_demand", 
          type = "quantitative"
          ),
        size = list(
          value = 60
          ) 
      )
    )
  )
))

Exercise 2

a

jobs <- jsonlite::fromJSON(
  "https://cdn.jsdelivr.net/npm/vega-datasets@2.8.0/data/jobs.json"
  )

jobs_perc <- jobs %>%
  group_by(job, year, sex) %>%
  summarise(total_perc = sum(perc))

`summarise()` has grouped output by 'job', 'year'. You can override using the
`.groups` argument.

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = jobs_perc),
  mark = "bar",
  encoding = list(
    x = list(
      field = "year", 
      type = "ordinal"
      ),
    y = list(
      field = "total_perc", 
      type = "quantitative"
      ),
    color = list(
      field = "sex", 
      type = "nominal"
      )
  )
))

The visualization helps us see how the proportion of men and women working in these occupations changed over time, which is what perc represents.

b

filtered_jobs <- jobs %>%
  filter(year %in% c(1850, 1900, 1950, 2000))

filtered_jobs_perc <- jobs_perc %>%
  filter(year %in% c(1850, 1900, 1950, 2000))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = filtered_jobs_perc),
  mark = "point",
  encoding = list(
    x = list(
      field = "job", 
      type = "nominal", 
      axis = list(labelAngle = 45)
      ),
    y = list(
      field = "total_perc", 
      type = "quantitative", 
      axis = list(title = "Percent (%)", format = ".0%")
      ),
    row = list(
      field = "year", 
      type = "ordinal"
      )
  )
))

c

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = filtered_jobs),
  mark = "point",
  encoding = list(
    x = list(
      field = "job", 
      type = "nominal", 
      axis = list(labelAngle = 45)
      ),
    y = list(
      field = "perc", 
      type = "quantitative", 
      axis = list(title = "Percent (%)", format = ".0%")
    ),
    color = list(
      field = "sex", 
      type = "nominal"
      ),
    row = list(
      field = "year", 
      type = "ordinal"
      )
  )
))

Exercise 3

gas_prices <- read.csv("https://calvin-data304.netlify.app/data/pump_price_for_gasoline_us_per_liter.csv")
gas_prices[,-1] <- lapply(gas_prices[,-1], as.numeric)

Warning in lapply(gas_prices[, -1], as.numeric): NAs introduced by coercion

gas_prices_long <- gas_prices %>%
  pivot_longer(
    cols = -country,
    names_to = "year",
    values_to = "price"
 ) %>%
  mutate(
    year = gsub("X", "", year),
 ) %>%
  drop_na()

codes <- read.csv("https://raw.githubusercontent.com/lukes/ISO-3166-Countries-with-Regional-Codes/master/all/all.csv")

combined <- left_join(gas_prices_long, codes, by = c("country" = "name"))

as_vegaspec(
  list(
  `$schema` = "https://vega.github.io/schema/vega-lite/v5.json",
  data = list(values = combined),
  mark = "line",
  encoding = list(
    x = list(
      field = "year", 
      type = "ordinal"
      ),
    y = list(
      field = "price", 
      type = "quantitative"
      ),
    color = list(
      field = "country", 
      type = "nominal",
      legend = NULL
      ),
    facet = list(
      field = "region", 
      type = "nominal",
      columns = 3
      )
  )
))